We can learn a lot about our presidential candidates from the individual words they use. Through computational analysis, we’ve pulled together lists of words that each candidate tends to use more frequently than their competition. These words shed light on the issues, policies, and rhetoric of individual candidates on an extremely granular level, ultimately giving us valuable insight into the people vying for the title of President of the United States.
We looked at transcripts from the fifteen main presidential debates that took place between August 6, 2015 and February 13, 2016—six Democratic Party debates and nine Republican Party debates. From these transcripts, we created six candidate files (Clinton, Cruz, Kasich, Rubio, Sanders, and Trump) that contained all the words spoken by individual candidates during the course of these debates. Using a corpus analysis tool, we compared the individual candidate files to the corpus of all the debate transcripts during this period. This allowed us to identify words that were used more frequently by an individual candidate than by the larger group. (Find more information on the statistics behind these calculations here.)
Below are the top twenty words favored by each candidate in our analysis.
We also looked at the words by party here.